We have several ePMP 4500 devices installed, and we are consistently observing that the downlink RSSI is significantly lower than the uplink RSSI across multiple units. For example, as shown in the attached screenshot, the RSSI for downlink/uplink is around -69/-59 dBm.
We can’t pinpoint the cause of this RSSI discrepancy. Has anyone experienced a similar issue with the ePMP 4500, or can anyone suggest potential reasons or settings to investigate?
Any assistance would be greatly appreciated!
This can be expected behavior, at least in the US. In UNII-1 & UNII-3 bands, FCC allows SM to operate as a PtP device, allowing much higher EIRP. I’m not sure of the rules in other parts of the world. Check the transmit power of the SM, listed a few lines down on the home page. I believe the max is 30 dBm. So EIRP in the uplink direction with a 425 could be 54. What is transmit power of the AP? US EIRP max in the downlink is 36. In this case downlink system gain is 36 dB EIRP out of AP + 24 dB antenna gain = 60 dB. In the uplink, it is 54 EIRP + 17 dB sector antenna gain = 71 dB. This doesn’t take into account MIMO gain. I’m not sure how the MIMO gain is incorporated into the RSSI readings on the radios.
All this being said, in our early testing with the 4500 AP to 425 we saw RSSI 6-8 dB below what we expected. Never got a great answer from Cambium. Have you checked expected RSSI using link planner or a basic RF link budget calculator?
Yes we sure have.
Interestingly we have some weird data that further mystifies.
We took a connectorized 5G unit and slapped it on a 27dB dish only to have the RSSI at both ends be exactly the same.
With the increased gain and transmit power, aways away from me automatic transmit power threshold, why would that be?
the difference you see on 4 or 8 chain radios is because we measure RSSI with management packets transmitted by 0,1 chains only. Data packets are transmitted with all chains with beamforming. It gives additional 6-9 dB.
We are working to make this indication less confusing.
Hi there. I wonder if you can explain how your comment is applicable to my situation.
Let me give you all the details.
We deployed a 5Ghz AP and linked several of the SM kits, we’re seeing downlinks in the realm of -60 - -65 or so.
As an experiment, we took a connectorized 5Ghz (400C) unit to replace one SM (425); initial stats were -67 downlink -58 uplink, after the 400C with increased transmit power and a 27dB dish, stats were again -67 and -58. We thought that was pretty odd.
Can you help me figure out how your statement is applicable to mine?
Thank you.
Hi, I think you meant to address this to me, and it would make sense in this context, but in that case, I should be seeing a good signal level on the client side (since the client-side equipment is 2x2, and the base station reports that the signal level in one direction is good). However, when I check the CPE, the signal level is even more disappointing. Based on simulations, the signal level should be much better. There’s a range extender on the 425, and a 0.6m Jirous antenna with 29dBi gain on the 400c.
Where we’ve upgraded 3000L to 4500L (same sector antenna, just changed radio) we’ve seen a 3-6db reduction in downlink RSSI, and modulation drop too. And that’s both the customers where we’ve upgraded their 300-25 up to a 4525, and also for customers that have their existing 300-25 connected to the 4500L. However, maybe this is just a reporting issue, or something to do with these APs reporting signals from the beacon frames, rather than the data or something? I don’t know.
I have also thought maybe at the same time we changed frequencies (which can affect signal) or channel width, or maybe we inadvertently move the antenna somewhat, or introduced a pigtail issue.
I’m not sure, we haven’t tracked it down yet, and we’re are kind of stuck and probably not going to upgrade much more to AX going forward anyway.
What transmit power are the SMs running? I’m not familiar with the ETSI power limits, but if there’s a difference in transmit power (not EIRP, conducted power) between the AP and SM, then there should normally be an equal difference in uplink/downlink signal levels.
from your description it sounds like a HW issue. It can be burn out due to thunder-strike or so.
Could you Raise Ticket in the top right corner of this page and attach techsupport files.
If we confirm HW we will proceed with RMA.
Just wanted to chime in and say it would be great to get signals from all the chains on the clients. I have mentioned this before on a past thread regarding the ePMP 4600 4x4. We recently installed a ePMP4500C with two 60 degree 4x4 horn arrays to provide a 100 degree coverage, with the center of the coverage getting the full benefit of the 8x8 beamforming as that is where most of our members are located.
Clients on the edge of the 60 degree 4x4 horn that is connected to chains 2,3/6,7 are far enough off that they barely pickup chains 0,1.
So far performance has been fantastic on a very dirty 80Mhz channel, however seeing a downlink -80 and uplink of -66 for clients on the one edge of the coverage has definitely thrown myself and our field tech for a loop. I have been judging connections based on modulations and speed tests, which has proven to be a decent indicator over RSSI and SNR.
This radio has been behaving like this from the very first minute. This case is not unique either. I opened a ticket for it and attached the file. Please take a look at it.
